Drilling oil/gas/geothermal wells has been done in a similar way for decades. Basically, a drilling fluid with a density high enough to counter balance the pressure of the fluids in the reservoir rock, is used inside the wellbore to avoid uncontrolled production of such fluids. However, in many situations, it can happen that the bottomhole pressure is reduced below the reservoir fluid pressure. At this moment, an influx of gas, oil, or water occurs, named a kick. If the kick is detected in the early stages, it is relatively simple and safe to circulate the invaded fluid out of the well. After the original situation is restored, the drilling activity can proceed. However, if, by any means, the detection of such a kick takes a long time, the situation can become out of control leading to a blowout. According to Skalle, P. and Podio, A. L. in “Trends extracted from 800 Gulf Coast blow-outs during 1960–1996” IADC/SPE 39354, Dallas, Tex., March 1998, nearly 0.16% of the kicks lead to a blowout, due to several causes, including equipment failures and human errors.
On the other hand, if the wellbore pressure is excessively high, it overcomes the fracture strength of the rock. In this case loss of drilling fluid to the formation is observed, causing potential danger due to the reduction in hydrostatic head inside the wellbore. This reduction can lead to a subsequent kick.
In the traditional drilling practice, the well is open to the atmosphere, and the drilling fluid pressure (static pressure plus dynamic pressure when the fluid is circulating) at the bottom of the hole is the sole factor for preventing the formation fluids from entering the well. This induced well pressure, which by default, is greater than the reservoir pressure causes a lot of damage, i.e., reduction of near wellbore permeability, through fluid loss to the formation, reducing the productivity of the reservoir in the majority of cases.
Since among the most dangerous events while drilling conventionally is to take a kick, there have been several methods, equipment, procedures, and techniques documented to detect a kick as early as possible. The easiest and most popular method is to compare the injection flow rate to the return flow rate. Disregarding the drilled cuttings and any loss of fluid to the formation, the return flow rate should be the same as the injected one. If there are any significant discrepancies, drilling is stopped to check if the well is flowing with the mud pumps off. If the well is flowing, the next action to take is to close the blow-out preventer equipment (BOP), check the pressures developed without circulation, and then circulate the kick out, adjusting the mud weight accordingly to prevent further influx. Some companies do not check flow if there is an indication that an influx may have occurred, closing the BOP as the first step.
This procedure takes time and increases the risk of blow-out, if the rig crew does not quickly suspect and react to the occurrence of a kick. Procedure to shut-in the well can fail at some point, and the kick can be suddenly out of control. In addition to the time spent to control the kicks and to adjust drilling parameters, the risk of a blow-out is significant when drilling conventionally, with the well open to the atmosphere at all times.
The patent literature includes several examples of methods for kick detection, including U.S. Pat. No. 4,733,233 (Grosso) which discloses a method for kick detection using a downhole device, known as an MWD, instead of detecting by fluid flow. An MWD measures gas kick only, by wave perturbations which are created ahead of the influx and detected. This method does not detect liquid (water or oil) kicks.
Among the methods available to quickly detect a kick the most recent ones are presented by Hutchinson, M and Rezmer-Cooper, I. in “Using Downhole Anular Pressure Measurements to Anticipate Drilling Problems”, SPE 49114, SPE Annual Technical Conference and Exhibition, New Orleans, La., 27–30 September, 1998. Measurement of different parameters, such as downhole annular pressure in conjunction with special control systems, adds more safety to the whole procedure. The paper discusses such important parameters as the influence of ECD (Equivalent Circulating Density, which is the hydrostatic pressure plus the friction losses while circulating the fluid, converted to equivalent mud density at the bottom of the well) on the annular pressure. It is also pointed out that if there is a tight margin between the pore pressure and fracture gradients, then annular pressure data can be used to make adjustments to mud weight. But, essentially, the drilling method is the conventional one, with some more parameters being recorded and controlled. Sometimes, calculations with these parameters are necessary to define the mud weight required to kill the well. However, annular pressure data recorded during kill operations have also revealed that conventional killing procedures do not always succeed in keeping the bottomhole pressure constant.
In some methods it is conventional to estimate pore pressure on detection of a kick in order to circulate the kick out of the well. U.S. Pat. No. 5,115,871 (McCann) discloses a method to estimate pore pressure while drilling by monitoring two parameters and monitoring respective change therein. GB 2 290 330 (Baroid Technology Inc) discloses a method of controlling drilling by estimating pore pressure from continually evaluated parameters, to take into account wear of drill bit.
Other publications deal with methods to circulate the kick out of the well. For example, U.S. Pat. No. 4,867,254 teaches a method of real time control of fluid influxes into an oil well from an underground formation during drilling. The injection pressure pi and return pressure pr and the flow rate Q of the drilling mud circulating in the well are measured. From the pressure and flow rate values, the value of the mass of gas Mg in the annulus is determined, and the changes in this value monitored in order to determine either a fresh gas entry into the annulus or a drilling mud loss into the formation being drilled.
U.S. Pat. No. 5,080,182 teaches a method of real time analysis and control of a fluid influx from an underground formation into a wellbore being drilled with a drill string while drilling and circulating from the surface down to the bottom of the hole into the drill string and flowing back to the surface in the annulus defined between the wall of the wellbore and the drill string, the method comprising the steps of shutting-in the well, when the influx is detected; measuring the inlet pressure Pi or outlet pressure Po of the drilling mud as a function of time at the surface; determining from the increase of the mud pressure measurement, the time tc corresponding to the minimum gradient in the increase of the mud pressure and controlling the well from the time tc.
U.S. Pat. No. 3,470,971 (Dower) and U.S. Pat. No. 5,070,949 (Gavignet) are further examples of kick circulation methods. Dower discloses an automated method for kick circulation, intended to keep wellbore pressure constant by adjusting back pressure by means of a choke during circulation. Gavignet discloses a method which comprises measuring gas in the annulus as the fluid influx travels upwards during circulation.
It is observed that in all the cited literature where the drilling method is the conventional one, the shut-in procedure is carried out in the same way. That is, literature methods are directed to the detection and correction of a problem (the kick), while there are no known methods directed to eliminating said problem, by changing or improving the conventional method of drilling wells. Thus, according to drilling methods cited in the literature, the kicks are merely controlled.
In the last 10 years, a new drilling technique, underbalanced drilling (UBD) is becoming more and more popular. This technique implies a concomitant production of the reservoir fluids while drilling the well. Special equipment has been developed to keep the well closed at all times, as the wellhead pressure in this case is not atmospheric, as in the traditional drilling method. Also, special separation equipment must be provided to properly separate the drilling fluid from the gas, and/or oil, and/or water and drilled cuttings.
EP 1 048 819 (Baker-Hughes) discloses an UBD method, and regulates injection of different fluid types to maintain a downhole pressure which ensures underbalance condition. U.S. Pat. No. 5,975,219 (Sprehe) is not as such designed as an UBD method, rather as a method which operates with a closed well head when drilling with a gas drilling fluid only, in order to contain the gas. However there are similarities to the UBD method.
The UBD technique has been developed initially to overcome severe problems faced while drilling, such as massive loss of circulation, stuck pipe due to differential pressure when drilling depleted reservoirs, as well as to increase the rate of penetration. In many situations, however, it will not be possible to drill a well in the underbalanced mode, e.g., in regions where to keep the wellbore walls stable a high pressure inside the wellbore is needed. In this case, if the wellbore pressure is reduced to low levels to allow production of fluids the wall collapses and drilling cannot proceed.
Accordingly, the present application relates to a new concept of drilling whereby a method and corresponding instrumentation allows that kicks may be detected early and controlled much quicker and safer or even eliminated/mitigated than in prior art methods.
Further, it should be noted that the present method operates with the well closed at all times. That is why it can be said that the method, herein disclosed and claimed, is much safer than conventional ones.
In wells with severe loss of circulation, there is no possibility to detect an influx by observing the return flow rate. Schubert, I. J. and Wright, J. C. in “Early kick detection through liquid level monitoring in the wellbore”, LADC/SPE 39400, Dallas, Tex., March 1998 propose a method of early detection of a kick through liquid level monitoring in the wellbore. Having the wellbore open to atmosphere, here again the immediate step after detecting a kick is to close the BOP and contain the well.
The excellent review of 800 blow-outs occurred in Alabama, Texas, Louisiana, Mississipi, and offshore in the Gulf of Mexico cited hereinbefore by Skalle, P. and Podio, A. L. in “Trends extracted from 800 Gulf Coast blow-outs during 1960–1996” IADC/SPE 39354, Dallas, Tex., March 1998 shows that the main cause of blow-outs is human error and equipment failure.
Nowadays, more and more oil exploration and production is moving towards challenging environments, such as deep and ultra-deepwater. Also, wells are now drilled in areas with increasing environmental and technical risks. In this context, one of the big problems today, in many locations, is the narrow margin between the pore pressure (pressure of the fluids—water, gas, or oil—inside the pores of the rock) and the fracture pressure of the formation (pressure that causes the rock to fracture). The well is designed based on these two curves, used to define the extent of the wellbore that can be left exposed, i.e., not cased off with pipe or other form of isolation, which prevents the direct transmission of fluid pressure to the formation. The period or interval between isolation implementation is known as a phase.
In some situations a collapse pressure (pressure that causes the wellbore wall to fall into the well) curve is the lower limit, rather than the pore pressure curve. But, for the sake of simplicity, just the two curves should be considered, the pore pressure and fracture pressure one. A phase of the well is defined by the maximum and minimum possible mud weight, considering the curves mentioned previously and some design criteria that varies among the operators, such as kick tolerance and tripping margin. In case of a kick of gas, the movement of the gas upward the well causes changes in the bottomhole pressure. The bottomhole pressure increases when the gas goes up with the well closed. Kick tolerance is the change in this bottomhole pressure for a certain volume of gas kick taken.
Tripping margin, on the other hand, is the value that the operators use to allow for pressure swab when tripping out of the hole, to change a bit, for example. In this situation, a reduction in bottomhole pressure, caused by the upward movement of the drill string can lead to an influx.
According to FIG. 1 attached, based on prior art designing of wells for drilling, typically a margin of 0.3 pound per gallon (ppg) is added to the pore pressure to allow a safety factor when stopping circulation of the fluid and subtracted from the fracture pressure, reducing even more the narrow margin, as shown by the dotted lines. Since the plot shown in FIG. 1 is always referenced to the static mud pressure, the compensation of 0.3 ppg allows for the dynamic effect while drilling also. The compensation varies from scenario to scenario but typically lies between 0.2 and 0.5 ppg.
From FIG. 1, it can be seen that the last phase of the well can only have a maximum length of 3,000 ft, since the mud weight at this point starts to fracture the rock, causing mud losses. If a lower mud weight is used, a kick will happen at the lower portion of the well. It is not difficult to imagine the problems created by drilling in a narrow margin, with the requirement of several casing strings, increasing tremendously the cost of the well. In some critical cases, a difference as small as 0.2 ppg is found between the pore and fracture pressures. Moreover, the current well design shown in FIG. 1 does not allow to reach the total depth required, since the bit size is continuously reduced to install the several casing strings needed. In most of these wells, drilling is interrupted to check if the well is flowing, and frequent mud losses are also encountered. In many cases wells need to be abandoned, leaving the operators with huge losses.
These problems are further compounded and complicated by the density variations caused by temperature changes along the wellbore, especially in deepwater wells. This can lead to significant problems, relative to the narrow margin, when wells are shut in to detect kicks/fluid losses. The cooling effect and subsequent density changes can modify the ECD due to the temperature effect on mud viscosity, and due to the density increase leading to further complications on resuming circulation. Thus using the conventional method for wells in ultra deep water is rapidly reaching technical limits.
On the contrary, in the present application the 0.3 ppg margins referred to in FIG. 1 are dispensed with during the planning of the well since the actual required values of pore and fracture pressures will be determined during drilling. Thus, the phase of the well can be further extended and consequently the number of casing strings required is greatly reduced, with significant savings. If the case of FIG. 1 is considered, the illustrated number of casings is 10, while by graphically applying the method of the invention this number is reduced to 6, according to FIG. 2 attached. This may be readily seen by considering only the solid lines of pore and fracture gradient to define the extent of each phase, rather than the dotted lines denoting the limits that are in conventional use. In order to overcome these problems, the industry has devoted a lot of time and resources to develop alternatives. Most of these alternatives deal with the dual-density concept, which implies a variable pressure profile along the well, making it possible to reduce the number of casing strings required. In some drilling scenarios, such as in areas where higher than normal pore pressure is found in deepwater locations, the dual density drilling system is the only one that may render the drilling economical.
The idea is to have a curved pressure profile, following the pore pressure curve. There are two basic options:                injection of a lower density fluid (oil, gas, liquid with hollow glass spheres) at some point for example WO 00/75477 (Exxon Mobil) which operates with injection of a gas phase lightweight fluid in a system having pressure control devices at the wellhead and at the seabed and detects changes in seabed pressure at the wellhead and compensates accordingly);        placement of a pump at the bottom of the sea to lift the fluid up to the surface installation for example WO 00/49172 (Hydril Co) which uses a choke to regulate the return flow and the well bore pressure to a pre-selected level.        
There are advantages and disadvantages of each system proposed above. The industry has mainly taken the direction of the second alternative, due to arguments that well control and understanding of two-phase flow complicates the whole drilling operation with gas injection.
Thus, according to the IADC/SPE 59160 paper “Reeled Pipe Technology for Deepwater Drilling Utilizing a Dual Gradient Mud System”, by P. Fontana and G. Sjoberg, it is possible to reduce casing strings required to achieve the final depth of the well by returning the drilling fluid to the vessel with the use of a subsea pumping system. The combination of seawater gradient at the mud line and drilling fluid in the wellbore results in a bottomhole equivalent density that can be increased as illustrated in FIG. 2 of the paper. The result is a greater depth for each casing string and reduction in total number of casing strings. It is alleged that larger casing can then be set in the producing formation and deeper overall well depths can be achieved. The mechanism used to create a dual gradient system is based on a pump located at the sea bottom.
However, there are several technical issues to be overcome with this option, which will delay field application for some years. The cost of such systems is also another negative aspect. Potential problems with subsea equipment will make any repair or problem turn into a long down-time for the rig, increasing even further the cost of exploration.
Another method currently under development by the industry is the injection of liquid slurry containing lightweight spheres at the bottom of the ocean, in the annulus, and injecting conventional fluid through the drillstring. The combination of the light slurry and the conventional fluid coming up the annulus creates a lighter fluid above the bottom of the ocean, and a denser fluid below the bottom of the ocean. This method creates also a dual-density gradient drilling or DGD. This alternative is much simpler than the expensive mud lift methods, but there are still some problems and limitations, such as the separation of the spheres from the liquid coming up the riser, so that they can be injected again at the bottom of the ocean. The slurry injected at the bottom of the ocean has a high concentration of spheres, whereas the drilling fluid being injected through the drillstring does not have any sphere, therefore the requirement for separation of the spheres at the surface.
One approach in DGD is currently being developed by Maurer Technology using oilfield mud pumps to pump hollow spheres to the seafloor and inject the lightweight spheres into the riser to reduce the density of the drilling mud in the riser to that of the seawater. It is alleged that the use of oilfield mud pumps instead of the subsea pumping DGD systems currently being developed will significantly reduce operational costs.
A safety requirement for offshore drilling with a floating drilling unit is to have inside the well, below the mud line, a drilling fluid having sufficient weight to balance the highest pore pressure of an exposed drilled section of the well. This requirement stems from the fact that an emergency disconnection might happen, and all of a sudden, the hydrostatic column provided by the mud inside the marine riser is abruptly lost. The pressure provided by the mud weight is suddenly replaced by seawater. If the weight of the fluid remaining inside the well after the disconnection of the riser is not high enough to balance the pore pressure of the exposed formations, a blowout might occur. This safety guard is called Riser Margin, and currently there are several wells being drilled without this Riser Margin, since there is no dual-density method commercially available so far.
There are three other main methods of closed system drilling: a) underbalanced flow drilling, which involves flowing fluids from the reservoir continuously into the wellbore is described and documented in the literature; b) mud-cap drilling, which involves continuous loss of drilling fluid to the formation, in which fluid can be overbalanced, balanced or underbalanced is also documented; c) air drilling, where air or other gas phase is used as the drilling fluid. These methods have limited application, i.e., underbalanced and air drilling are limited to formations with stable wellbores, and there are significant equipment and procedural limitations in handling produced effluent from the wellbore. The underbalanced method is used for limited sections of the wellbore, typically the reservoir section. This limited application makes it a specialist alternative to conventional drilling under the right conditions and design criteria. Air drilling is limited to dry formations due to its limited capability to handle fluid influxes. Similarly Mud-Cap drilling is limited to specific reservoir sections (typically highly fractured vugular carbonates).
Thus, the open literature is extremely rich in pointing out methods for detecting kicks, and then methods for circulating kicks out of the wellbore. Generally all references teach methods that operate under conventional drilling conditions, that is, with the well being open to the atmosphere. However, there is no suggestion nor description of a modified drilling method and system, which, by operating with the well closed, controlling the flow rates in and out of the wellbore, and adjusting the pressure inside the wellbore as required, causing that influxes (kicks) and fluid losses do not occur or are extremely minimized, such method and system being described and claimed in the present application. In a particular advantage of the present invention the system and method differ from UBD methods which operate with closed well but generate a constant controlled influx of fluid, as hereinbefore described. Moreover the system and method are adapted for operation with a substantially incompressible drilling fluid whereby changes in pressure/flow may be detected or made at the wellhead and the effect downhole may be accurately calculated without complex pressure differential considerations. Nevertheless for offshore drilling, the present method and system employing back pressures can also be used with lightweight fluids so that the equivalent drilling fluid weight above the mud line can be set lower than the equivalent fluid weight inside the wellbore, with increasing safety and low cost relative to drilling with conventional fluids.